Results from prospective skin paint studies of eight chemicals tested in the Tg.AC transgenic mouse model were compared with the results for the same chemicals tested in the two-year rodent bioassay. There was an accordance between the two assays for four of them. These results also indicated that the Tg.AC model was more likely to detect genotoxic or nongenotoxic carcinogens that were trans-species carcinogens in the two-year bioassay and would be unlikely to detect nongenotoxic carcinogens that induced only liver or kidney tumors in a single sex/species. The investigation to detect key early inflammatory or growth-control response genes that modify or associate with skin-tumor induction in Tg.AC is continuing. TGF-alpha, a gene thought to be required for cell proliferation, was found to be dispensable for skin tumorigenesis in Tg.AC mice, but it is highly likely that other ligands that activate the EGF receptor were replacing the requirement for TGF-alpha. Induction of expression of GM-CSF, a very early inflammatory response gene appears to be a requirement for TPA induction of tumors. Recently several novel VEGF/VPF transcripts were identified in TPA induced skin tumors of Tg.AC mice and their role in tumor induction is being investigated. These studies are relevant to human health because ras gene mutations are found in over 40% of all human tumors and the identification of genes that control key signal transduction pathways that modify or inhibit oncogenic ras gene expression can lead to insights for cancer prevention.